from bs4 import BeautifulSoup
import re
import urllib.parse
import requests
import os


def writeImg(imgUrl):

    imgReply = requests.get(imgUrl)  # stream=True,
    # print(r.headers)  # r.text,  r.history, r.text,lastChild
    with open('{}'.format(os.path.basename(urllib.parse.urlparse(imgUrl).path)), "ab") as file:
        file.write(imgReply.content)

xml = open(r'E:\LOFTER\LOFTER.xml', encoding="utf8").read()
soup = BeautifulSoup(xml, 'xml')  # 必须以xml形式解析，否则不能识别photoLinks tag


imgs = soup.find_all('img', src=True)  # 有的img没有src
for img in imgs:
    # print(img['src'])
    writeImg(img['src'])


photoLinksTags = soup.findAll('photoLinks')
# #re.findall(r'"orign": ".+?"', xml)
# print(eval(imgs[-2].string))  # [0] imgs[-2].string,
for photoLinksTag in photoLinksTags:
    for photoItem in eval(photoLinksTag.string):  # photoLinksTag里包含一系列photoItems，以dict方式展现
        # print(photoItem["orign"])
        writeImg(photoItem["orign"])